OTHER PROJECT INFORMATION - Project Summary/Abstract Regular economic activity exposes workers to unexpected shocks to their earnings, employment, and career development. For example, more than 10% of American workers are laid off every year. While for many workers these shocks may be short-lived, a growing literature shows that short-term career shocks - such as involuntary job losses - can have large and persistent effects for certain groups of workers. For example, mature displaced workers tend to experience long-term earnings losses, reduced employment probabilities, early retirement, increased job instability, as well as reductions in consumption and a loss in health insurance coverage. A long literature has demonstrated a strong correlation between earnings and employment and health. Moreover, a growing literature link stress from economic uncertainty and unhappiness to health outcomes. However, at present no comprehensive study of the effect of job loss, economic outcomes, and short and long-term health exists, partly due to a lack of data on detailed longitudinal career and health information for a large sample of workers. This is a particular concern for public policy, since many job losers may be eligible for disability insurance and most eventually become eligible for Medicare. Thus, an important part of the cost of job loss is borne by society through higher health expenditures. We propose to study the effects of job losses during mass-layoffs and firm closings on short and long term health outcomes using several large administrative data sets accessed through the Social Security Administration through an ongoing cooperation. The data contains information on earnings, disability, death, and Medicare expenditures and is ideal for this purpose, since it provides longitudinal information on earnings and health outcomes and employers for a 10% sample of Americans ranging from 1978 to 2005. We also use detailed health-related and demographic data from the Survey of Income and Program Participation available for a sub-sample of our data. The data allows us to isolate events at the firm level triggering job loss that are not correlated with workers' prior health status and earnings. Thereby, we can isolate situations in which workers are affected by truly external negative events, and follow the effect on earnings, entry into disability insurance, and mortality over 15 to 20 years after a job loss. As workers receive disability insurance or turn 65, we can also analyze the effect of job loss on Medicare expenditures. Our research design allows us to (1) evaluate the effects of job loss during mass-layoffs or firm closings on short- and long-term health outcomes and health expenditures; (2) to study the channels through which job loss affects health outcomes and health expenditures; and (3) to assess the social costs of job loss and earnings dispersion in terms of increased expenditures through Medicare and Social Security Disability Insurance. In response to reviewers' comments, we also use the survey data to assess workers' health prior to job loss, to study health effects of job loss other than mortality, and to directly examine key channels such as presence of health insurance or spousal income.